Bees have amazing cognitive capabilities. The scouts explore and memorize the surrounding landscape and build up a highly detailed neuronal representation of their environment. Once back in the hive, locations are communicated to others by translating location information into body movements; a behavior known as the honeybee waggle dance. We investigated the bee dance in previous experiments with a honeybee robot and noticed that foragers exhibit preferences for certain nestmates in the process of decoding the dance. Do bees form stable “peer groups” throughout their lives? Do bees prefer dances that point to known locations or that offer known food qualities? How much previous experience plays a role in the process of becoming a dance-follower?
In the past, answering those questions was very laborious. Biologists would sit in front of an observation hive, record videos and review these videos manually over many weeks or months. We use the various tools Computer Science offers to develop a system which allows tracking every single individual inside the hive over their entire lifespan. We can study which bees are communicating with whom and which experiences in the past might have facilitated this interaction.
We are happy to work with our partner, the North-German Supercomputing Alliance at the Zuse Institut Berlin where all our data (~700 Terabyte) is stored and analyzed on a supercomputer.
Team
Tim Landgraf, Benjamin Wild, David Dormagen, Fernando Wario, Leon Sixt, Franziska Boehnisch, Dominik Dreiner, Sascha Wissmann, Andreas Berg, Jens Hagemeister
Former student members (thank you!): Simon Wichmann, Balduin Laubisch, Amjad Saadeh, Christian Tietz, Mareike Ziese, Alexander Rau, Benjamin Rosemann, Alexa Schlegel, Kadir Tugan, Jonas Piotrowski, Sascha Witte, Maria Sparenberg, Franziska Lojewski, Lukas Kairies
Funding
The North-German Supercomputing Alliance (HLRN) granted us storage and computing resources to process massive amounts of image data.
Latest posts in BeesBook
Ben and Leon @ ICLR 2017 (4/26/2017) Posted in: BeesBook, Machine Learning, News, Talks and Presentations - Solving problems in Machine Learning very often relates to the question how to represent the data. While in the past most problems Read More
Tracking System Improved! (9/13/2016) Posted in: BeesBook, General, News - Our master student Jakob Mischek just finished his work on our tracking system which just got way better. Check out Read More
New Image Processing Pipeline Finished! (9/12/2016) Posted in: BeesBook, General, News, Uncategorized - Our team member Leon Sixt has developed a new image processing component for the decoding of bee markers. The method Read More
BeesBook: Live Hive Life! (7/15/2016) Posted in: Announcement, BeesBook, General, News - Get a drink and a bowl of popcorn. As of now you can watch our bees live! We are launching Read More
Methods-Paper on BeesBook-System Published (9/25/2015) Posted in: BeesBook, General, New Paper, News, Publication - We just published the first tracking system for honeybee colonies in Frontiers in Ecology and Evolution. Thanks to the HLRN Read More
New article on the honeybee dance and BeesBook (3/27/2015) Posted in: BeesBook, General, News, Project in the Media - Raúl Rojas frequently publishes articles on AI and robotics or science in general in Telepolis, a popular German online magazine. Read More
Supercomputing proposal for BeesBook accepted! (9/16/2014) Posted in: Announcement, BeesBook, General, News - We have successfully applied to The North-German Supercomputing Alliance (HLRN) for computing and storage resources. The HLRN grants more than half Read More
Tag Team Action! (7/23/2014) Posted in: BeesBook, General, News - We have marked all bees in our observation hive and the first full-fledged recording season has begun. We are very Read More
RoboBee and BeesBook on Deutsche Welle TV (7/7/2014) Posted in: BeesBook, General, Project in the Media, RoboBee - 
Tracking Bees with New Tag Design! (2/13/2014) Posted in: BeesBook, General, News - We changed the design of the binary tags to a circular code. Look below!